Every tutoring company lists the benefits of math tutoring, and the lists are the same: confidence, better grades, personalised attention. They are true as far as they go. What a parent needs is different: which benefits are real, how big they are, what has to be true for them to happen, and when tutoring does not help at all. Here is the version with the evidence.
Why math, in particular
Math is cumulative in a way most subjects are not. A child who missed a history unit can still follow the next one. A child who did not master fractions cannot do ratios, or algebra, or slopes, or probability. The gap does not stay the same size; everything taught afterwards is built on it. A class teacher with thirty students has to move on. That is the reason one-to-one teaching helps more in math than almost anywhere else, and it is also why the first job of a math tutor is diagnosis.
Eight benefits
1. The real gap is found. A child stuck on Grade 7 equations is often missing something from Grade 4: negative numbers, fractions, the meaning of the equals sign. A tutor who tests backwards finds it in one or two sessions. This is the benefit the rest depend on.
2. Someone watches the working. In class a teacher sees answers. A tutor sees each line as it is written and stops the mistake as it happens: the dropped minus sign, the fraction flipped the wrong way. Feedback within seconds is far more effective than a mark the next day.
3. The pace is the child's. Faster through what is known, slower on what is not, and no moving on until it is secure.
4. Confidence, earned. Not from praise, but from getting ten in a row right on something that used to be impossible. Math anxiety falls when competence rises, in that order.
5. Homework stops being a fight. Partly because the child can do it, and partly because the explanation now comes from someone who is not a parent.
6. Exam technique. Showing working for method marks, checking, managing time, reading what a question asks. These are taught skills, worth a grade on their own in high school.
7. Stretch for a child who is ahead. Tutoring is not only for catching up. A strong student who is bored can work a year ahead or on competition problems.
8. A second adult who knows the child's math. Someone who can tell a parent exactly what is and is not understood, and can tell the school the same.
How large is the effect?
| Source | What it found |
|---|---|
| Nickow, Oreopoulos and Quan (2020), a review of 96 randomised trials of tutoring | A pooled effect of about 0.37 standard deviations on learning: large by the standards of education research |
| Education Endowment Foundation toolkit | One-to-one tuition: about five months of extra progress on average; small-group tuition about four |
| University of Chicago Education Lab, studies of daily in-school math tutoring for Chicago ninth and tenth graders | Math test scores up by about 0.16 to 0.37 standard deviations in one year, and fewer course failures |
In plain terms: tutoring is one of the best-evidenced ways to raise achievement that researchers have studied. The averages hide a wide range, and what separates the good results from the weak ones is known.
What makes math tutoring work
- Frequency. Two or more sessions a week. The strongest studies are of tutoring three to five times a week.
- The same tutor. The first month is spent learning the child. A changing tutor starts again each time.
- Diagnosis first. A tutor who begins with this week's homework, without checking what lies beneath it, treats the symptom.
- Connection to the school's course. Tutoring that follows the child's own curriculum and textbook is used the next morning. A separate program is not.
A fifth, quieter one: the child does the writing. A session in which the tutor explains for forty minutes feels helpful and teaches little.
When math tutoring does not help
- When it is homework completion. If the tutor's hand is doing the work, the grade for homework rises and the test grade does not.
- When it is occasional. One session before each test is reassurance, not teaching.
- When the problem is not math. A child who cannot read the word problems, cannot see the board, is not sleeping, or has unaddressed attention or anxiety difficulties needs that looked at first.
- When the child is doing fine. Tutoring a child with no gap and no goal buys supervision.
- When the tutor cannot teach the level. Above Algebra 2, subject depth matters; a tutor one chapter ahead of the student is not enough. The kinds of tutor and what each is for are in different types of tutors.
Who gains most
| The child | What tutoring does |
|---|---|
| Behind by a year or more | The largest measured gains, with frequent sessions |
| "Understands in class, fails the test" | Finds the skill that is only partly secure; adds checking and exam technique |
| Anxious about math | Small daily successes; a place to be wrong without an audience |
| Moving between schools, states or countries | Maps what was covered against what the new school expects |
| Ahead and bored | Acceleration, kept in step with the school so nothing is skipped |
How to tell within a month
After four weeks of regular sessions you should be able to answer yes to most of these:
- The tutor has named, specifically, what was missing.
- Your child can show you one thing they could not do a month ago.
- Homework takes less time.
- Your child goes to the session without a fight.
- You receive something after each session: a note, the marked work, a recording.
Grades take longer, six to twelve weeks, because they depend on when the next test falls. If none of the five is true after a month, change the arrangement, not the child. How long sessions should be at each age is in how long should a tutoring session be, and what it costs in how much does a math tutor cost per hour.
How we do it
One child, one tutor with a master's degree in mathematics, the same tutor every week, on the child's own school course. The first class is a diagnosis; the plan after it names the gaps. The child writes on a shared whiteboard and the tutor reads every line. After each class the parent gets the recording, the notes and the marked homework. US$10 to $13 an hour, billed monthly, with no package, which makes two sessions a week affordable, and two a week is what the research supports. The free 30-minute first class is a real lesson.
Questions parents ask
1What are the benefits of math tutoring?
2Does math tutoring actually work?
3Is a math tutor worth the money?
4How often should a child have math tutoring?
5At what age is math tutoring most useful?
See how we teach this, 1-on-1 online β
Shobha
Founder and lead tutor, Science with Shobha
Teaching maths, science and English to children online since 2018. 500+ students, 20+ tutors, families in nine countries.
